Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A computer program product comprising a non-transitory machine-readable medium storing instructions which, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations comprising: receiving a first medical image of a patient taken at a first time and at a first position and a second medical image of the patient taken at a second time and at a second position, the first medical image and the second medical image different due to a deformation present in the second medical image; comparing the first medical image and the second medical image using a phase correlation method, the comparing comprising: calculating an inverse Fourier transform of a cross-power spectrum of the first medical image and the second medical image, the inverse Fourier transform having a peak spread around neighboring voxels due to the deformation; and calculating a displacement of the peak based at least partially on the intensities of the neighboring voxels; determining, based on the displacement, a change to at least one of a physical location and a physical orientation of the patient; and correcting, based on the determined change, the second position of the patient to more closely conform to the first position of the patient.
Medical imaging analysis. This invention addresses the problem of accurately determining and correcting for patient movement or deformation between medical image acquisitions. It describes a computer program product stored on a non-transitory medium. When executed by a processor, these instructions enable the system to receive two medical images of a patient, acquired at different times and potentially different positions. These images are understood to differ due to deformation in the second image. The core operation involves comparing these two images using a phase correlation method. This comparison includes calculating the inverse Fourier transform of their cross-power spectrum. A key aspect is that this inverse Fourier transform will exhibit a peak spread across neighboring voxels due to the deformation. The system then calculates the displacement of this peak, taking into account the intensities of the surrounding voxels. Based on this calculated displacement, the system determines a change in the patient's physical location and/or orientation. Finally, using this determined change, the system corrects the second position of the patient so that it more closely matches the first position.
2. A computer program product as in claim 1 , wherein the inverse Fourier transform of the cross-power spectrum is normalized.
The invention relates to signal processing, specifically improving the accuracy of cross-power spectrum analysis in systems where signals are corrupted by noise or interference. The cross-power spectrum is a mathematical tool used to analyze the relationship between two signals in the frequency domain, but its accuracy can be degraded by noise. The invention addresses this by applying an inverse Fourier transform to the cross-power spectrum and then normalizing the result. Normalization ensures that the processed signal maintains consistent amplitude levels, reducing distortions caused by noise or varying signal strengths. This technique is particularly useful in applications such as radar, sonar, and communication systems, where precise signal analysis is critical. The normalization step compensates for variations in signal power, improving the reliability of frequency-domain measurements. By normalizing the inverse Fourier transform of the cross-power spectrum, the invention enhances the signal-to-noise ratio and provides a more accurate representation of the underlying signal relationships. This method is applicable to any system where cross-power spectrum analysis is used to extract meaningful information from noisy signals.
3. A computer program product as in claim 1 , the comparing further comprising: finding a maximum intensity of the inverse Fourier transform of the cross-power spectrum; and selecting as the peak, from a plurality of voxels having intensities greater than a threshold determined based on the maximum intensity, a voxel for which a sum of voxel intensities of neighboring voxels around the voxel is highest, wherein the neighboring voxels are in a window sized as a fraction of a number of voxels along each dimension of a common registration grid.
This invention relates to image processing, specifically to improving the accuracy of peak detection in medical imaging, such as in positron emission tomography (PET) or magnetic resonance imaging (MRI). The problem addressed is the challenge of precisely identifying peaks in image data, which is critical for tasks like tumor localization or functional brain mapping. Existing methods may struggle with noise or overlapping signals, leading to inaccurate peak detection. The invention describes a method for refining peak detection by analyzing the inverse Fourier transform of a cross-power spectrum. First, the maximum intensity of the inverse Fourier transform is identified. Then, from a set of voxels (3D pixels) with intensities above a dynamically determined threshold (based on the maximum intensity), the method selects the peak voxel. The selection is based on the sum of intensities of neighboring voxels within a predefined window, where the window size is a fraction of the total voxels along each dimension of a common registration grid. This ensures that the chosen peak is not only locally strong but also supported by its surrounding context, improving robustness against noise and artifacts. The approach enhances the reliability of peak detection in medical imaging applications.
4. A computer program product as in claim 3 , wherein the threshold is 0.9 times the maximum intensity.
A system and method for analyzing signal intensity in a data processing environment involves determining a threshold value for identifying significant signal peaks. The system processes input data to detect signal peaks and compares their intensities against a dynamically calculated threshold. The threshold is set at 0.9 times the maximum intensity observed in the data, ensuring that only the most prominent peaks are selected for further analysis. This approach filters out noise and minor fluctuations, improving the accuracy of signal detection. The method includes steps for normalizing the input data, identifying local maxima, and applying the threshold to retain only the highest-intensity peaks. The system may be used in applications such as sensor data analysis, medical imaging, or communication signal processing, where distinguishing relevant signals from background noise is critical. By dynamically adjusting the threshold based on the maximum observed intensity, the system adapts to varying signal conditions, enhancing reliability in real-world scenarios. The invention provides a robust solution for signal peak detection, reducing false positives and improving decision-making in automated data analysis systems.
5. A computer program product as in claim 3 , wherein the fraction is 0.05.
A system and method for optimizing data processing in a distributed computing environment addresses inefficiencies in resource allocation and task scheduling. The invention focuses on improving computational efficiency by dynamically adjusting the fraction of data processed in parallel across multiple nodes. Specifically, the system calculates an optimal fraction of data to distribute for parallel processing, balancing workload distribution and minimizing overhead. The fraction is determined based on factors such as network latency, node capacity, and task complexity. In one embodiment, the fraction is set to 0.05, meaning only 5% of the total data is processed in parallel while the remaining 95% is handled sequentially or in a different configuration. This approach reduces communication overhead between nodes, lowers latency, and enhances overall system throughput. The method includes monitoring system performance metrics, adjusting the fraction dynamically, and redistributing tasks accordingly. The invention is particularly useful in large-scale data processing applications, such as big data analytics and machine learning, where efficient resource utilization is critical. By optimizing the fraction of parallel processing, the system achieves better performance without requiring significant hardware upgrades.
6. A computer program product as in claim 3 , the comparing further comprising determining the displacement of the peak by calculating a centroid of the voxel intensities of the neighboring voxels in the window.
This invention relates to medical imaging analysis, specifically a method for detecting and quantifying displacement of anatomical features in volumetric imaging data, such as CT or MRI scans. The problem addressed is accurately identifying small shifts in tissue structures over time or between different imaging sessions, which is critical for monitoring disease progression, treatment response, or structural changes in organs. The invention involves analyzing a three-dimensional image dataset composed of voxels (volumetric pixels) to detect displacement of a peak intensity region within a defined window of neighboring voxels. The key innovation is determining the displacement by calculating the centroid of the voxel intensities in the window. This centroid-based approach provides a precise measurement of the peak's position, improving accuracy over simpler methods that rely on single-voxel comparisons or thresholding. The process includes selecting a window around a peak intensity region in the image, analyzing the voxel intensities within that window, and computing the centroid to determine the peak's displacement. This technique is particularly useful for tracking subtle changes in anatomical structures, such as tumor margins, organ boundaries, or vascular features, where small shifts may indicate significant clinical developments. The method enhances the reliability of longitudinal imaging studies by reducing errors from noise or partial volume effects. The invention is implemented as a computer program product, enabling integration into existing medical imaging software for automated analysis.
7. A computer program product as in claim 1 , the correcting further comprising causing translating or rotating of a patient by causing movement of a patient couch or bed.
This invention relates to medical imaging systems, specifically addressing the challenge of correcting patient positioning errors during imaging or treatment procedures. The system includes a computer program product that enhances image-guided medical procedures by automatically adjusting the position of a patient to ensure accurate alignment with the imaging or treatment device. The correction process involves translating or rotating the patient by controlling the movement of a patient couch or bed. This adjustment compensates for misalignment detected during the procedure, ensuring precise targeting and reducing the need for manual repositioning. The system integrates with imaging devices to detect positional discrepancies and calculates the necessary adjustments to optimize patient alignment. By automating these corrections, the invention improves procedural accuracy, reduces radiation exposure, and enhances patient safety. The solution is particularly useful in radiation therapy, where precise patient positioning is critical for effective treatment delivery. The invention may also include additional features such as real-time feedback and adaptive correction algorithms to further refine positioning accuracy.
8. A computer program product as in claim 7 , wherein the first medical image and the second medical image are aligned by generating one or more commands to automatically translate or rotate the patient couch or bed.
This invention relates to medical imaging systems, specifically improving image alignment between multiple scans of a patient. The problem addressed is ensuring accurate registration of medical images taken at different times or from different angles, which is critical for diagnostic accuracy and treatment planning. Misalignment can lead to errors in diagnosis or therapy delivery. The invention involves a computer program product that processes medical images to detect misalignment between a first and a second image of the same patient. Once misalignment is detected, the system generates commands to automatically adjust the position of the patient couch or bed. This adjustment compensates for the misalignment by translating (moving linearly) or rotating the couch to realign the patient's position relative to the imaging system. The system may use image processing techniques to determine the necessary adjustments, ensuring that subsequent images are properly aligned with prior scans. This automation reduces manual intervention, improves efficiency, and enhances the reliability of medical imaging workflows. The invention is particularly useful in applications like radiation therapy, where precise alignment between planning and treatment images is essential.
9. A system comprising: at least one programmable processor; and a non-transitory machine-readable medium storing instructions which, when executed by the at least one programmable processor, cause the at least one programmable processor to perform operations comprising: receiving a first medical image of a patient taken at a first time and at a first position and a second medical image of the patient taken at a second time and at a second position, the first medical image and the second medical image different due to a deformation present in the second medical image; comparing the first medical image and the second medical image using a phase correlation method, the comparing comprising: calculating an inverse Fourier transform of a cross-power spectrum of the first medical image and the second medical image, the inverse Fourier transform having a peak spread around neighboring voxels due to the deformation; and calculating a displacement of the peak based at least partially on the intensities of the neighboring voxels; determining, based on the displacement, a change to at least one of a physical location and a physical orientation of the patient; and correcting, based on the determined change, the second position of the patient to more closely conform to the first position of the patient.
This system addresses the challenge of aligning medical images taken at different times or positions, where deformations or patient movement cause misalignment. The system uses a programmable processor and a non-transitory machine-readable medium storing instructions to process medical images. The system receives a first medical image taken at a first time and position and a second medical image taken at a later time and a different position, where the second image is deformed due to patient movement or other factors. To align the images, the system compares them using a phase correlation method. This involves calculating the inverse Fourier transform of the cross-power spectrum of the two images, which produces a peak spread across neighboring voxels due to deformation. The system then calculates the displacement of this peak by analyzing the intensities of the neighboring voxels. Based on this displacement, the system determines changes in the patient's physical location or orientation. Finally, the system corrects the second position of the patient to better match the first position, improving image alignment for accurate medical analysis. The method ensures precise registration of medical images despite deformations or movement.
10. A system as in claim 9 , wherein the inverse Fourier transform of the cross-power spectrum is normalized.
The invention relates to signal processing systems, specifically those involving Fourier transforms and cross-power spectrum analysis. The system addresses the challenge of accurately reconstructing signals from their cross-power spectrum representation, particularly when normalization is required to ensure proper scaling and interpretation of the results. The system includes a processing unit configured to compute the cross-power spectrum of two input signals, which represents the correlation between their frequency components. The processing unit then applies an inverse Fourier transform to the cross-power spectrum to reconstruct the time-domain signal. A key feature of the system is that the inverse Fourier transform result is normalized, ensuring that the output signal is properly scaled and free from artifacts that could arise from unnormalized transformations. This normalization step is critical for applications where precise amplitude or phase information is required, such as in communications, radar, or audio processing. The system may also include additional components, such as filters or amplifiers, to preprocess the input signals before cross-power spectrum computation or to post-process the normalized output. The normalization step may involve dividing the inverse Fourier transform result by a scaling factor derived from the cross-power spectrum or other statistical properties of the input signals. This ensures that the reconstructed signal maintains consistent amplitude characteristics regardless of the input signal strengths. The invention improves the reliability and accuracy of signal reconstruction in systems where cross-power spectrum analysis is employed.
11. A system as in claim 9 , the comparing further comprising: finding a maximum intensity of the inverse Fourier transform of the cross-power spectrum; and selecting as the peak, from a plurality of voxels having intensities greater than a threshold determined based on the maximum intensity, a voxel for which a sum of voxel intensities of neighboring voxels around the voxel is highest, wherein the neighboring voxels are in a window sized as a fraction of a number of voxels along each dimension of a common registration grid.
This invention relates to image processing, specifically to a method for identifying a peak in a cross-power spectrum derived from medical imaging data, such as ultrasound or MRI. The problem addressed is accurately locating a peak in a cross-power spectrum to improve image registration or alignment, which is critical for medical diagnostics and treatment planning. The system processes a cross-power spectrum obtained from two or more medical images. It first computes an inverse Fourier transform of the cross-power spectrum to generate a spatial representation. The system then identifies the maximum intensity in this transformed data. A threshold is set based on this maximum intensity to filter out low-intensity voxels. From the remaining voxels, the system selects the peak voxel where the sum of intensities of neighboring voxels within a defined window is highest. The window size is a fraction of the total voxels along each dimension of a common registration grid, ensuring adaptability to different image resolutions. This approach enhances peak detection by considering both local intensity and neighborhood context, improving the robustness of image alignment in medical imaging applications. The method is particularly useful in scenarios where noise or artifacts may obscure the true peak in the cross-power spectrum.
12. A system as in claim 11 , wherein the threshold is 0.9 times the maximum intensity.
A system for analyzing signal intensity in a measurement process involves determining a threshold value to identify significant signal events. The system processes input signals to detect peaks or other notable features, then compares these features against a dynamically calculated threshold. The threshold is set at 0.9 times the maximum intensity observed in the signal, ensuring that only the most prominent features are selected. This approach filters out noise and minor variations, improving the accuracy of signal analysis. The system may include preprocessing steps to condition the input signal, such as filtering or normalization, before applying the threshold. The thresholding mechanism can be applied in various domains, including sensor data analysis, communication systems, or medical signal processing, where distinguishing relevant signal components from background noise is critical. By using a fraction of the maximum intensity, the system adapts to varying signal strengths while maintaining consistent performance across different operating conditions. The method ensures reliable detection of significant events without manual adjustment, enhancing automation and reducing false positives.
13. A system as in claim 11 , wherein the fraction is 0.05.
A system for optimizing resource allocation in a distributed computing environment addresses inefficiencies in workload distribution across multiple processing nodes. The system dynamically adjusts the fraction of computational tasks assigned to a secondary processing node to balance load and improve performance. The fraction is set to 0.05, meaning 5% of tasks are directed to the secondary node while the remaining 95% are processed by the primary node. This configuration ensures that the primary node handles the majority of workloads, reducing latency and overhead associated with task redistribution. The secondary node serves as a backup or auxiliary processor, handling a controlled subset of tasks to prevent overloading the primary node while maintaining system responsiveness. The system monitors performance metrics such as task completion time and resource utilization to dynamically adjust the fraction if needed, ensuring optimal distribution under varying workload conditions. This approach enhances scalability and reliability in distributed computing environments by efficiently utilizing available processing resources.
14. A system as in claim 11 , the comparing further comprising determining the displacement of the peak by calculating a centroid of the voxel intensities of the neighboring voxels in the window.
The system relates to medical imaging analysis, specifically for detecting and quantifying anatomical or pathological changes in volumetric data, such as CT or MRI scans. The problem addressed is the accurate and automated detection of small shifts or displacements in anatomical features over time or between different imaging modalities, which is critical for monitoring disease progression, treatment response, or structural changes. The system processes volumetric image data, where each data point is represented as a voxel with an intensity value. A window of neighboring voxels is analyzed to identify a peak intensity, which corresponds to a feature of interest, such as a lesion or anatomical landmark. The system then determines the displacement of this peak by calculating the centroid of the voxel intensities within the window. The centroid provides a precise measure of the peak's position, accounting for variations in intensity distribution. This method improves accuracy in tracking small movements or changes in the feature's location, which is essential for clinical decision-making. The system may also include preprocessing steps to enhance image quality, such as noise reduction or normalization, and may apply the displacement calculation across multiple windows to generate a comprehensive map of changes. The centroid-based approach ensures robustness against noise and partial volume effects, making it suitable for high-resolution medical imaging applications.
15. A system as in claim 9 , the correcting further comprising causing translating or rotating of a patient by causing movement of a patient couch or bed.
A system for medical imaging or radiation therapy includes a patient couch or bed that can be automatically adjusted to correct misalignment between a patient and an imaging or treatment device. The system detects positional discrepancies between the patient and the intended treatment or imaging target, then compensates by translating or rotating the patient couch or bed. This ensures precise alignment, improving accuracy in procedures such as radiation therapy or diagnostic imaging. The system may use sensors or imaging feedback to determine the required adjustments, which are then executed by motors or actuators controlling the couch or bed. The couch or bed may move in multiple axes (e.g., lateral, longitudinal, vertical) or rotate to achieve the correct positioning. This automated correction reduces manual intervention, minimizes errors, and enhances treatment or imaging efficiency. The system may integrate with existing medical devices, such as linear accelerators or MRI machines, to provide real-time alignment adjustments during procedures. The couch or bed may also include features like locking mechanisms to secure the patient during movement. This technology addresses the challenge of maintaining precise patient positioning in dynamic medical environments, where slight misalignments can affect treatment efficacy or diagnostic accuracy.
16. A system as in claim 15 , wherein the first medical image and the second medical image are aligned by generating one or more commands to automatically translate or rotate the patient couch or bed.
This invention relates to medical imaging systems that align multiple medical images of a patient by adjusting the patient's position. The problem addressed is the need for precise alignment of medical images taken at different times or from different angles to ensure accurate diagnosis and treatment planning. Misalignment can lead to errors in medical assessments, such as in radiation therapy or surgical planning. The system includes a patient couch or bed that can be automatically adjusted in position. The system generates commands to translate (move linearly) or rotate the couch or bed to align a first medical image with a second medical image. This ensures that the images are properly registered, meaning they are spatially aligned for accurate comparison or analysis. The alignment process may involve image processing techniques to determine the necessary adjustments, which are then executed by actuators or motors controlling the couch or bed. This automation reduces manual intervention, improving efficiency and accuracy in medical imaging workflows. The system may also include feedback mechanisms to verify alignment and make further adjustments if needed. The invention is particularly useful in applications where precise image registration is critical, such as in radiation therapy, where misalignment could lead to incorrect treatment delivery.
17. A computer program product as in claim 1 , wherein the first medical image and the second medical image are received from at least one of an MRI imaging device or a CT imaging device.
This invention relates to medical imaging systems that process and analyze images from MRI or CT imaging devices. The technology addresses the challenge of accurately comparing and aligning medical images from different sources or modalities to improve diagnostic accuracy and treatment planning. The system receives a first medical image and a second medical image, where these images may originate from either an MRI (Magnetic Resonance Imaging) device or a CT (Computed Tomography) device. The images are processed to enhance their quality, align them spatially, and extract relevant anatomical or pathological features. The system may also include techniques for noise reduction, contrast enhancement, and feature detection to improve the reliability of the analysis. By integrating images from different modalities, the system enables more comprehensive and precise medical assessments, aiding in the detection and monitoring of diseases such as tumors, fractures, or other abnormalities. The invention enhances the diagnostic workflow by providing a unified platform for comparing and interpreting medical images from multiple sources, reducing the need for manual adjustments and improving efficiency in clinical settings.
18. A system as in claim 9 , wherein the first medical image and the second medical image are received from at least one of an MRI imaging device or a CT imaging device.
This invention relates to a medical imaging system designed to process and analyze medical images from different imaging modalities. The system addresses the challenge of integrating and comparing medical images obtained from different sources, such as MRI (Magnetic Resonance Imaging) or CT (Computed Tomography) devices, to improve diagnostic accuracy and workflow efficiency. The system receives a first medical image and a second medical image from at least one of an MRI imaging device or a CT imaging device. These images may be acquired from the same or different imaging modalities, allowing for cross-modal analysis. The system processes these images to align, register, or compare them, enabling healthcare professionals to assess anatomical or pathological features more comprehensively. The integration of images from different modalities helps overcome limitations inherent to single-modality imaging, such as MRI's soft-tissue contrast versus CT's bone and density visualization. The system may include additional components, such as image processing algorithms, user interfaces for visualization, and storage for archiving and retrieving images. By supporting multiple imaging modalities, the system enhances diagnostic capabilities, reduces redundancy, and improves patient care by providing a more holistic view of medical conditions. The invention is particularly useful in clinical settings where multiple imaging techniques are employed for accurate diagnosis and treatment planning.
Unknown
May 12, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.